Clinical evaluation of the optic nerve head

There are several practical consequences of discarding intraocular pressure (IOP) as central to the definition of glaucoma. One is that the clinician must become proficient at examining the optic nerve head (ONH) to appreciate the often subtle signs of glaucomatous optic atrophy; neither tonometry nor perimetry alone can be relied on to determine the presence of the disease. In fact, there is increasing evidence that alterations in the ONH are the earliest signs of primary open-angle glaucoma (POAG), and that visual field studies are more useful later in the disease process.

Similarly, instead of using the IOP for classification into ‘normal-tension’ or ‘high-pressure’ glaucoma, various subtypes of glaucomatous disease are postulated based on various appearances of the glaucomatous ONH, with specific disc changes often seen in constellation with associated clinical findings. Although technological advances have been made in imaging and quantifying the three-dimensional features of the ONH, the fact remains that the clinician’s mastery of discriminating ophthalmoscopy is indispensable for the appropriate management of the glaucoma patient.


The observer’s ability to stereoscopically evaluate the ONH with sufficient magnification is the essence of optic nerve surveillance in glaucoma. This can effectively be done at the slit lamp, using a variety of contact and non-contact lenses. It can also be performed using stereoscopically obtained disc photographs. Comparable information can be obtained from a variety of commercial imaging devices (see Ch. 14 ).

Slit-lamp funduscopy provides a good stereoscopic, direct view of the ONH when the pupil is at least 4 mm in diameter. Direct visualization of an upright image can be performed with a non-contact lens such as the Hruby or high-diopter (78D or 90D) fundus lens, or with contact devices such as the Zeiss four-mirror or Goldmann macular lens. With proper calibration and attention to detail, reliable measures of the optic disc diameter and disc area can be generated with such techniques, whose data are comparable to laborious planimetric measurements. For smaller pupils, a contact lens is often required for stereopsis. Indirect and inverted disc visualization with the 60D, 78D, or 90D fundus lenses can best be obtained when the pupil is dilated. The consistent advantage of high-magnification stereoscopic viewing is that many subtle alterations in the ONH can be detected, such as discrepancies between the cup size based on color criteria and contour criteria. Similarly, the shallow cupping of myopia is more obvious with a narrowed slit-lamp beam, and the disc often can be better seen this way in patients with early cataracts.

Monocular examination of the ONH is done with a hand-held, direct ophthalmoscope. The ease of this method makes it suitable for glaucoma screening and for interval evaluations that seek information on specific disc findings, such as the presence or resolution of a disc hemorrhage. Direct ophthalmoscopy is best thought of as an adjunct to the stereoscopic evaluation, the latter providing the specific three-dimensional details that are then monitored monocularly by parallax viewing and creation of shadows. The halogen bulb in the direct ophthalmoscope provides a brighter view than standard bulbs; when used with a red-free green filter, the nerve fiber layer can be visualized effectively.

If the pupil is small, if the cornea is irregular, or if the eye moves (as in children or patients with nystagmus), the patient can be placed supine, and a smooth-domed Koeppe lens can be applied. This will hold the lids open and help steady the globe, both allowing gonioscopy and providing a clear (but minified) view of the posterior fundus and ONH with the direct ophthalmoscope.

Photographs of the ONH continue to remain an extremely useful technique for documenting change in the disc over time ( Fig. 13-1A, B ). Precisely because photographic slides are portable, durable and independent of constantly changing technological platforms, they conserve invaluable information for long-term care. For example, they are useful to obtain as a baseline before refractive corneal surgery in young myopes at risk for glaucoma, since the images of their discs allow future comparisons decades hence. Baseline photographs should be taken in glaucoma suspects and glaucoma patients at the time of the initial visit, and then at intervals of every 6–18 months, depending on the patient’s stage of disease and clinical stability. Careful stereo evaluation of these pictures, using commercial viewers or +10 lenses, allows for the appreciation of subtle changes in the contour of the cup and shape of the neuroretinal rim (NRR), changes in the pathway of vessels, subtle disc hemorrhages not clinically appreciated or alterations in the peripapillary choroid ( Figs. 13-2 and 13-3 ).

Fig. 13-1

(A, B) Fundus photographs of a left eye six months apart showing increasing disc cupping during period of uncontrolled intraocular pressure (upper 30s). Note rather uniform enlargement of the disc cup. (C, D) Hand-drawn diagrams, as noted in patient’s chart, of the same progressive cupping seen in photographs. The accompanying cross indicates estimations of maximal vertical and horizontal dimensions of the cup.

Fig. 13-2

Optic disc pit. The inferior pole of this optic disc has a horizontally elliptical greyish defect into which the inferior vein falls. This is a dramatic optic pit. This eye had a corresponding dense superior visual field defect.

Fig. 13-3

(A) The initial photograph of the right optic disc in a patient with primary open-angle glaucoma. (B) The patient after a 12-year interval. During this interval, there has been concentric enlargement of the cup. An area of focal thinning of the neural rim can be observed near a cilioretinal vessel in the superotemporal region. The bent portion of this vessel could not be visualized well before the loss of the overlying neural rim tissue.

From Campbell DG, Netland PA: Stereo atlas of glaucoma, St Louis, Mosby, 1998.

The clinician’s disc drawings are a useful adjunct to disc photographs and should be performed regularly on all patients (see Fig. 13-1C, D ). They are valuable for two reasons: they require the clinician to pay attention to subtle details in the ONH, and they are an incentive to regularly review previous drawings and photographs to assess disc stability. Likewise, they can potentially be as valuable as disc photographs in determining progression. Various drawing routines have been devised, but attention to stereoscopic details – such as the vertical and horizontal demarcation of the cup; the integrity and regularity of the NRR; the configuration of vessels at the disc margins; the appearance of laminar pores or disc hemorrhages; and peripapillary disc changes – can be methodically delineated when attention is given to their diagrammatic rendering.

Spaeth and co-workers have proposed an elaborate but reproducible clinical scheme for diagramming and staging the extent of glaucomatous disc damage, using slit-lamp and direct ophthalmoscopic technology. The Disc Damage Likelihood Scale (DDLS) distinguishes 10 stages of progressive glaucomatous changes of the disc, whose clinical significance is discriminated based on whether the disc size is small, average or large. Classification is abetted by a standardized chart with examples ( Fig. 13-4 ). First, either a 60D, 66D or 90D lens is used at the slit lamp to estimate the disc size in millimeters with a reticule, and the value multiplied by lens-power-dependent constants: this determines a small, average or large optic nerve size. Next, the neuroretinal rim is assessed at its narrowest point (i.e., the cup axis is discriminated at its largest extent) by direct ophthalmosocopic exam, and notated; the disc is drawn, with careful attention to the NRR. Lastly the DDLS is invoked by integrating the disc diagram, the narrowest rim-to-disc ratio and the nomogram; assume an average size, and adjust afterwards. Two sequential parameters are encountered with progressive disease: the radial width of the NRR at its narrowest point; and in areas of total NRR loss, the circumferential extent (in degrees) of absent rim tissue. The score is determined for each eye, and the quantitative values assigned have high inter- and intra-observer agreement, as well as good correlation with visual field loss.

Fig. 13-4

The Disk Damage Likelihood Scale (DDLS) is based on the radial width of the neuroretinal rim measured at its thinnest point. The unit of measurement is the rim/disk ratio, that is, the radial width of the rim compared to the diameter of the disk in the same axis. When there is no rim remaining, the rim/disk ratio is 0. The circumferential extent of rim absence (0 rim/disk ratio) is measured in degrees. Caution must be taken to differentiate the actual absence of rim from sloping of the rim as, for example, can occur temporally in some patients with myopia. A sloping rim is not an absent rim. Because rim width is a function of disk size, disk size must be evaluated prior to attributing a DDLS stage. For small disks (diameter <1.50 mm) the DDLS stage should be increased by 1; for large disks (diameter >2.00 mm) the DDLS stage should be decreased by 1. This is done with a 60D to 90D lens with appropriate corrective factors. The Volk 66D lens minimally underestimates the disk size. Corrective factors for other lenses are: Volk 60D × .88, 78D × 1.1, 90D × 1.33. Nikon 60D × 1.03, 90D × 1.63.


In large part because of the prodigious efforts by Jonas and co-workers in meticulously delineating the morphometric and pathogenic details of the ONH in glaucoma and optic atrophy, there is a framework for approaching the massive literature on the features of glaucomatous changes in the ONH derived from clinical observations, investigational imaging, and histopathologic correlations. Eight intrapapillary findings and four associated peripapillary features address the spectrum of glaucomatous alterations of the ONH. Following this, proposed subclassifications of glaucoma based on patterns of ONH appearance are discussed.


With meticulous stereoscopic observation of the intrinsic structures of the optic papilla, the clinician can discern two aspects of the optic disc (its size and shape); two aspects of the NRR (its size and shape); three aspects of the optic cup (its size and configuration, and the cup:disc ratio); and the relative position of the central retinal vascular trunk and its branches to the laminar surface and disc architecture.

Optic disc size

The mean area of the optic disc in whites is 2.1–2.8 mm 2 and is independent of age after the first decade of life. Disc size is related to race – it tends to be smaller in whites, intermediate among Asians, and largest among blacks. There are several helpful clinical tools in estimating disc size during routine examination. When the smallest 5° aperture of the Welch Allyn direct ophthalmoscope is projected onto the retina in eyes with a refractive error within ±4D of emmetropia, the circular spot has an area of 1.7 mm 2 – and thus provides a reliable estimation of whether an optic disc is clinically larger or smaller than normal ( Fig. 13-5 ). Another technique for clinical-research use is the calculation of the disc area based on measuring the horizontal and vertical diameters of the ONH. After adjusting the slit-lamp beam in the vertical (V) and horizontal (H) meridians and multiplying the obtained lengths by a factor of 1.26, these values are then entered into the modified formula for an ellipse:

<SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='Area=π/4×(H×V)in  mm2.’>Area=π/4×(H×V)in  mm2.Area=π/4×(H×V)in  mm2.
Area = π / 4 × ( H × V ) in  mm 2 .

Figure 13.5

Drawing of the light spot projected onto the retina when the fundus is visualized with the small (5°) aperture of the Welch Allyn direct ophthalmoscope. The light spot is normally slightly smaller than the average normal optic disc and provides a convenient clinical technique for estimating optic disc size for eyes within ±4D of emmetropia.

From Airaksinen PJ, Tuulonen A, Werner EB: Clinical evaluation of the optic disc and retinal nerve fiber layer. In: Ritch R, Shields MB, Krupin T, editors: The glaucomas, 2nd edn, St Louis, Mosby, 1996.

Formulas for using different sizes of indirect lenses at the slit lamp to measure the disc have also been generated.

Within the refractive range of −5 to +5D, the optic disc size shows little significant variation. Hyperopes over +5 show smaller optic discs, and high myopes have larger optic discs. Jonas has characterized macrodiscs , manifesting large areas (over 4.2 mm 2 ), as either primary or secondary. Primary macrodiscs are independent of age and refraction and can be asymptomatic (presumably hereditary) or symptomatic, as in the morning glory syndrome or congenital optic pit. Secondary macrodiscs include the enlarging ONHs seen with progressive myopia or with uncontrolled congenital glaucoma.

Optic disc size correlates with a variety of morphometric and clinical features. Small discs are associated with ONH drusen, pseudopapilledema, and non-arteritic anterior ischemic optic neuropathy. Large discs demonstrate greater neuroretinal rim area, more axonal fibers, larger size and number of laminar pores, and other expressions of increased ocular cellular components, such as more retinal pigment epithelial cells and photoreceptors, cilioretinal arteries, and so on.

It is intriguing that intermediate disc size is associated with arteritic anterior ischemic optic neuropathy, central retinal vein occlusion, and the most common forms of glaucoma – POAG, juvenile open-angle glaucoma, age-related (‘senile sclerotic’) open-angle glaucoma, and pseudoexfoliation open-angle glaucoma. Logically, one might deduce that smaller discs, with relatively smaller axonal numbers (reduced axonal reserves) and a higher ratio of laminar pores to disc area, might be more susceptible to damage. (In fact, slight enlargement of cupping in small discs is apt to be clinically significant.) Conversely, one might argue that larger discs, despite their greater axonal numbers (reserves?), sustain greater laminar displacement and are subject to a larger area of laminar pressure gradients. Compensatory factors, such as comparable connective tissue proportions and pore size distribution in the lamina cribrosa (as seen in black eyes when compared with white eyes), may play a role in minimizing the effect of optic disc size on the cascade of glaucomatous insults in the lamina cribrosa.

Optic disc shape

The oval form of the normal optic disc has a vertical dimension approximately 7–10% larger than the horizontal dimension. The shape shows no correlation with age, sex, body weight, or height – it only correlates with the extent of corneal astigmatism. This relationship is particularly seen with high degrees of astigmatism, with the axis corresponding to the longest axis of the ONH. Recognition of this fundus finding in children may alert the clinician to evaluate the corneas and forestall refractive amblyopia.

In myopes of less than 8D, there is no apparent difference in the optic disc shape from normals. Nor is there any correlation between the optic disc shape and the NRR area, perimetric defects, or the susceptibility to glaucoma. However, with increasing myopia, and especially in eyes over 12D myopic, there is increasing ovality of the disc. This suggests stretching or tractional vectors are not evenly distributed across the myopic optic disc and that may play a role in the pathogenesis of glaucomatous optic atrophy in these eyes.

Neuroretinal rim size (NRR)

The NRR is the intrapapillary extension of the nerve fiber layer and is hence a critical parameter for evaluation. The direct correlation with the optic disc size – the larger the disc, the larger the NRR – is reflected in a direct correlation with axonal count and the area and number of laminar pores. This indicates that there is a greater axonal reserve capacity in eyes with larger optic discs. Nevertheless, there is a great deal of inter-individual variability of the NRR size, depending on specific factors such as axonal counts, axonal densities within the ONH, variations in laminar architecture, and glial cell counts within the disc.

Neuroretinal rim shape

The characteristic vertical oval shape of the optic disc and the horizontal oval shape of the optic cup contribute to the NRR shape. In descending order, the rim is broadest in the inferior disc, then the superior disc, then the nasal disc, and thinnest in the temporal disc. This asymmetry may reflect in part the fact that the center of the ONH is 0.53±0.34 mm horizontally above the foveola. Hence a relatively larger number of axons exit the eye through the relatively more cramped inferior portion of the ONH – which would account for the optimal visibility of the retinal nerve fiber layer (RNFL) in the inferotemporal aspect of the disc.

This gradient of axonal distribution, represented by the variable shape of the NRR in different disc quadrants, also correlates with the morphology of the lamina cribrosa – the largest pores and the least amount of interpore connective tissue are in the inferior and superior poles, compared with the nasal and temporal sectors. Similarly, the thinnest axons are temporal, and the thickest are in the polar regions. These findings all suggest that differential axonal populations distributed in a structurally non-homogeneous lamina cribrosa may well account for asymmetric cupping of the glaucomatous disc and characteristic visual field loss patterns of the disease.

In progressive glaucoma, the NRR is diffusely lost in all sectors, and its dimunition is predictive for subsequent visual field loss. Neuroretinal rim loss manifests earliest in the inferotemporal and superotemporal regions, followed by the temporal sector, and lastly the nasal rim. This is virtually identical to the patterns of visual field damage seen clinically. Also of clinical interest is the observation that the sector of the NRR farthest from the central trunk of retinal vessels may be affected by rim loss earlier than other sectors.

Although some have advocated the evaluation of NRR pallor as a distinctive indicator of early glaucomatous damage, other studies indicate that this color-assessment variable provides little additional information over discrimination of the NRR area itself. With few exceptions, non-glaucomatous optic atrophy manifests neither with enlargement of the optic cup nor with a decrease in the NRR area – but rather with increased NRR pallor and attenuation of the peripapillary retinal vessels.

Optic cup size in relation to optic disc size

The larger the optic disc, the larger the optic cup. As with macrodiscs, macrocups have been described and subclassified into primary types (physiologic and fixed after the first year of life) and secondary types (because of either increasing myopic enlargement of the disc or progressive glaucomatous atrophy).

Special attention must be paid to small optic discs, which usually have virtually no cup. Hence the earliest signs of glaucomatous progression may be more easily appreciated by evaluating the RNFL or peripapillary choroid. Even the development of a small cup in an eye previously without any cupping may indicate early damage in the crowded disc.

Optic cup configuration and depth

There is no standard pattern for the development of glaucomatous cupping of the ONH ( Fig. 13-6 ). This cupping may begin as a symmetric enlargement of the physiologic cup, but usually some portion of the rim erodes more rapidly than the rest. In myopic eyes and discs with age-related open-angle glaucoma, the cup tends to be shallow, especially temporally. In other kinds of glaucoma, NRR thinning can be localized and appear as a notch or, less frequently, as a pit at the disc rim. If notches are present both inferiorly and superiorly, the cup becomes vertically oval like a football.

Fig. 13-6

(A) Optic disc with slight physiologic cupping, cup:disc ratio (C/D) 0.4. (B) Optic disc with moderate cupping, C/D 0.5. (C) Optic disc with moderate cupping, C/D 0.8. (D) Optic disc with severe cupping, C/D 0.9. (E) Cupping of lower portion of the optic disc, C/D 0.5–0.6. (F) Slightly cupped myopic disc with peripapillary atrophy. Contact lens ophthalmoscopy demonstrates cupping that extends to the periphery of the disc.

The slope of the wall of the disc cup generally is steepest nasally and becomes shallowest temporally, with the upper pole having a somewhat steeper slope than the lower pole.

Cupping occurs slowly unless the pressure is very high. We have seen patients with pressure in the 40s develop an increase of 0.2–0.3 in cup:disc ratio in a matter of weeks. In young children and infants, the elastic capacities of the infant nerve allow transient and reversible disc cupping to be produced by pressure on the globe during examination of the patient under anesthesia.

Reversal of glaucomatous disc cupping can be dramatic in young patients following surgical or medical lowering of IOP ( Fig. 13-6 ). Reversal is less dramatic in older patients, presumably because reduced elasticity of the scleral tissue in older patients does not allow the cup to resume its prior configuration. Also, cupping that results from actual loss of nerve tissue is unlikely to reverse.

Cup:disc ratios

Because of the vertically oval optic disc and the horizontally oval optic cup, cup:disc ratios are usually larger horizontally than vertically in normal eyes; in only 7% of normals is this pattern reversed. This is in stark contrast to eyes manifesting early glaucomatous loss, in which the vertical cup:disc ratio increases faster than the horizontal ratio. Appreciation of increasing vertical cupping is clinically very useful.

Because the cup:disc ratio depends on the highly variable optic disc and optic cup diameters, normal ratios span from 0.0 to 0.8. They are larger in eyes with large optic discs and smaller in eyes with small optic discs. Stereoscopic viewing tends to yield higher estimations of the cup:disc axes than monocular viewing with a direct ophthalmoscope.

When drawing or noting the cup:disc ratio, it is particularly useful to indicate both the maximal horizontal (H) and the maximal vertical (V) dimensions. Often these are written in one-tenth units (e.g., .5 H & .6 V’); carrying out the estimation one decimal place further (e.g., ‘.55 H & .65 V’) indicates an intermediate estimation rather than the viewer’s hyperacuity precision (see Fig. 13-1C and D ). Estimations by the same observer may vary by 0.1–0.2 units over time; hence accurate drawings indicating precise landmarks, or stereophotographs, can greatly supplement the clinical record.

To a limited extent, cupping in adult glaucoma is reversible, but not nearly to the extent seen in children (see Fig. 13-7 ). Compressive lesions of the extrabulbar portion of the optic nerve can cause profound visual field loss that may recover dramatically when the compression is relieved. In glaucoma, this phenomenon exists to a small degree, in that apparent recovery of visual field loss has occurred following treatment of the glaucoma. Spaeth and co-workers suggest this as one way of recognizing the adequacy of treatment. In our experience, however, this recovery is rarely prominent and is certainly not manifest when defects in the nerve fiber layer have already appeared. As the axons die, they occupy less space in the scleral canal, and the cup enlarges. Quigley found up to 40% of some nerves’ axonal mass could be lost without having any recognizable field defect on Goldmann perimetry. Thus nerve damage can occur and progress with little or no field defect. Despite advances in computerized perimetry and other psychophysical tests for early glaucoma, progressive optic disc cupping without visual field loss remains an early indicator of glaucoma, assuming no other disease process is occurring.

Fig. 13-7

This 5-year-old child underwent two goniotomies. (A) The optic nerve before the first goniotomy. (B) The optic nerve before the second goniotomy. Note an increasing thickness of the rim with an increasing pinkness of the nerve head as pressure is lowered after each goniotomy.

Position of central retinal vessels and branches

It has been observed that there is local susceptibility to NRR loss in the rim sector farthest from the major trunk of the central retinal vessels. This specific relationship of glaucomatous loss can be useful to monitor in circumstances with an unusually shaped NRR.

As the cup enlarges, the retinal vessels, which usually pass perpendicularly through the disc tissue to reach the retina, are displaced externally following the receding nasal wall of the cup. Where the vessels shift from a more vertical orientation along the cup wall to a horizontal orientation on the retinal surface, there is a bend in the vessel. Change in the shape or position of that bend, as may be seen when comparing serial photographs, is a sensitive indicator of disc change and can be monitored with disc imaging ( Fig. 13-8 ).

Fig. 13-8

(A) Extension of the cup (notching) inferotemporally. (B) Seven years later; note thinning of the neural rim and shifts of the positions of blood vessels.

From Campbell DG, Netland PA: Stereo atlas of glaucoma, St Louis, Mosby, 1998.

Vessels that pass circumferentially across the temporal aspect of the cup have been called circumlinear vessels. If they pass the exposed depths of the cup, they are ‘bared.’ Baring of circumlinear vessels is seen because as the cup recedes, it exposes the vessels. Baring of circumlinear vessels is seen commonly in glaucomatous cups, but it may be seen in normal cups as well.


By paying attention to alterations in the area immediately surrounding the optic disc, valuable information can be determined about the glaucoma status of the ONH. Four phenomena should be evaluated – optic disc splinter hemorrhages; changes in the RNFL; variations in the diameter of retinal arterioles; and patterns of peripapillary choroidal atrophy (PPCA).

Optic disc hemorrhages

Nearly 30 years ago, Drance and co-workers revived interest in splinter hemorrhages on the ONH in glaucoma patients. These are either flame-shaped or blot hemorrhages that can occur at any location around the disc rim ( Fig. 13-9 ). They usually are located within the nerve fiber layer extending across the disc rim into the retina, but they may occur deep in the disc tissue. They last for a variable interval between 2 and 35 weeks. Because they appear in areas of preserved NRR, they are not usually seen in advanced cases of cupping in which little rim remains.

Fig. 13-9

MDisc hemorrhage at the inferior pole of the right optic disc.

From Campbell DG, Netland PA: Stereo atlas of glaucoma, St Louis, 1998, Mosby.

The literature on optic disc hemorrhage has been dominated by case series and case-controlled studies in eyes under surveillance for glaucoma. As a result, many assertions about the association of disc hemorrhage with one type of glaucoma (e.g., ‘low-tension glaucoma’) or about the specificity of this finding for predicting glaucomatous loss sometimes reflect this selection bias. A comprehensive Australian population-based study, using subjects rather than eyes, provides a more broad-based context for assigning meaning to disc hemorrhages.

This epidemiologic survey of mostly whites in the Blue Mountains near Sydney used a thorough glaucoma assessment including computerized fields and disc photographs on all participants. An overall prevalence rate of optic disc hemorrhage of 1.4% was found; this was slightly higher than the rates of 0.8% and 0.9% reported in the only two other population-based studies in the literature. Positive correlation was seen with increasing age and among women; no correlation was identified with a history of vascular events, smoking, aspirin use, or myopia. Most remarkable was that 70% of all disc hemorrhages were seen in subjects without any definite signs of glaucoma. Only 1 out of 4 patients over 50 years old with a disc hemorrhage demonstrated other disc and visual field signs of glaucoma. Thus the specificity of this sign as a screening tool does not seem particularly good, as reported elsewhere. Non-glaucomatous factors, such as aspirin use and diabetes, are also associated with such hemorrhages.

Nevertheless, among patients known to have glaucoma in this Australian study, disc hemorrhages were decidedly more common (13.8%); for every high-pressure glaucoma eye there were three eyes with ‘low-pressure’ disease. And when compared with normals, patients with ocular hypertension had twice the frequency of disc hemorrhage. Eyes with lower IOPs following filtration surgery showed fewer disc hemorrhages than pre-operatively. Others have demonstrated a strong association between disc hemorrhages in glaucomatous eyes with RNFL loss (especially inferotemporally), NRR notching, and discrete visual field loss. Such hemorrhages may precede progression of the disease, with subsequent disc and field changes manifesting between 1 and 7 years later.

Another long-term prospective evaluation, the Ocular Hypertensive Treatment Study, made several important observations about disc hemorrhages in selected patients with elevated IOPs. Of note was that only 16% of disc hemorrhages detected on stereophotographs were identified on funduscopic examination. It is of cautionary significance that so many of us clinicians are missing these subtle findings in the vast majority of cases. Another important finding was that though their patients were considered at risk for POAG at the time of their recruitment, after 7 years’ follow-up, some 86% of eyes that developed a disc hemorrhage did not manifest POAG according to strict, pre-defined criteria.

Others also report that there may be neither accelerated progression nor apparent functional loss associated with disc hemorrhage. Interestingly the morphometric para-meters of disc size, shape, NRR, peripapillary atrophy (and IOP or visual field loss) show no difference between eyes with unilateral disc hemorrhage. Asian eyes requiring filtration surgery for both open-angle and angle-closure glaucoma showed similar rates of disc hemorrhages as reported in white patients.

Although some of the data are contradictory, a cautionary value can be assigned to this ophthalmic finding. The identification of a disc hemorrhage in any eye compels considerations that glaucoma may be present. In an eye with known glaucoma, such a finding merits increased attention and surveillance.

Though some have proffered an association of disc hemorrhage and specific types of glaucoma – notably the focal type of normal-pressure glaucoma – others have noted the sampling problem bias of these assessments and claim that disc hemorrhages can be found in all types of glaucoma. This controversy reflects our lack of knowledge regarding the pathogenic mechanisms underlying the disc hemorrhage. For example, a higher IOP may stop a hemorrhage earlier, limit its size, and enhance its reabsorption faster than lower IOP. Hence the alleged connection between disc hemorrhage and ‘low-pressure glaucoma’ may simply reflect persistence of a more obvious bleed, leading to its greater clinical visibility.

Explanations that disc hemorrhages reflect ischemic events in the ONH contributing to glaucomatous progression are unconvincing on several grounds. Axonal damage occurs within the lamina cribrosa, and not in the NRR; and the retinal circulation that accounts for the capillary bed responsible for disc hemorrhages is not implicated in glaucomatous disease (see Ch. 12 ). Perhaps most telling is that disc hemorrhages are never associated with cotton-wool spots of the associated nerve fiber layer, as seen in focal ischemic infarction of the retina. On the other hand, explanations that say mechanical shearing of small disc vessels occurs with structural collapse of the progressively cupped disc are not supported by observations of eyes that have sustained sudden IOP elevations from ocular contusion yet do not manifest disc hemorrhages. Both the mechanism of disc hemorrhage and the precise vessels involved – arterioles, veins, or retinal radial peripapillary capillaries – remain unresolved.

Other possible causes of disc hemorrhages include anterior ischemic optic neuropathy, optic nerve drusen, central or branch retinal vein occlusion, diabetic retinopathy, vasculitis, papilledema, anticoagulation therapy, and posterior vitreous detachment. Most of these causes can be recognized by accompanying signs such as disc swelling, more diffuse hemorrhages throughout the retina, or vasculopathy. Posterior vitreous detachment can be diagnosed by recognizing the condensed vitreous ring that is torn away from its attachment at the ONH.

Nerve fiber layer defects

Ophthalmoscopic visible defects in the RNFL, first described by Hoyt and co-workers, represent visible loss of optic nerve axons from any form of optic atrophy. Two patterns of nerve fiber loss have been described – localized wedge defects and diffuse loss – which alone or in combination can be recognized in glaucoma patients. Localized loss is more easily and consistently recognized, but is less common.

Detection of RNFL defects is particularly useful in determining whether a glaucoma suspect has or has not yet manifested the structural changes of early glaucoma. Such RNFL changes can precede the appearance of functional changes in the visual field and thus have great value as early indicators of disease.

Clinical evaluation can be done by a variety of methods: red-free direct ophthalmoscopy, or a wide-angle camera with either blue or green filters and high-resolution black-and-white film such as Kodak Panatonic X. A photographic survey of the ONH region provides a particularly sensitive method of evaluation that affords more precise and studied assessment than a clinical exam alone of the patient. Specialized imaging techniques for assessing the RNFL have also been developed, such as computer-imaged height measures of the peripapillary nerve fiber layer, scanning laser polarimetry, photogrammetric measurements of RNFL thickness, and optical coherence tomography (see Ch. 14 ). The clinician should learn the basics of RNFL assessment either by using the slit lamp with a high-plus non-contact or contact lens, or by using red-free direct ophthalmoscopy. The red-free (green) filter combined with precise focusing enhances the appearance of the RNFL. The nerve fibers are seen most easily in the retinal area adjacent to the superior and inferior temporal aspects of the disc (superior and inferior Bjerrum’s area), where the RNFL is thickest. Here the fibers are closely packed and can be recognized by the bright linear reflexes reflected by the bundles.

The RNFL is more obvious in young people with moderately dark fundi and becomes increasingly difficult to see in older patients and in patients with media opacities or lighter fundus pigmentation. Beyond two disc-diameters, distance from the disc, the fibers diverge, and the RNFL thins, leaving normal darker spaces between the bundles. These slit-like spaces must not be confused with true localized slit RNFL defects, which are broader and darker, and extend up to the disc rim.

Localized RNFL defects appear as dark bands that fan outward from or near the disc margin, following the pattern of the nerve fiber layer. Localized loss may occur in conjunction with diffuse or generalized loss. Diffuse loss is more difficult to assess. To evaluate RNFL loss, the examiner must be familiar with the variations in appearance of the normal RNFL and photographic artifacts that can simulate diffuse RNFL loss. Comparing the RNFL appearance of one eye with the fellow eye can be helpful.

Careful observation of the vessels and vascular reflexes can help differentiate a poorly visible RNFL in a normal patient from a thinned or generally atrophic RNFL in a patient with disease. If the retinal vessel reflexes are bright and sharp and smaller branches are easily visualized, generalized atrophy may be present ( Fig. 13-10 ).

Feb 12, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Clinical evaluation of the optic nerve head

Full access? Get Clinical Tree

Get Clinical Tree app for offline access